AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915436472303616 |
|---|---|
| author | Lv, Changze Zhou, Jiang Long, Siyu Wang, Lihao Feng, Jiangtao Xue, Dongyu Pei, Yu Wang, Hao Zhang, Zherui Cai, Yuchen Gao, Zhiqiang Ma, Ziyuan Hu, Jiakai Gao, Chaochen Gong, Jingjing Song, Yuxuan Zhang, Shuyi Zheng, Xiaoqing Xiong, Deyi Bai, Lei Ouyang, Wanli Zhang, Ya-Qin Ma, Wei-Ying Zhou, Bowen Zhou, Hao |
| author_facet | Lv, Changze Zhou, Jiang Long, Siyu Wang, Lihao Feng, Jiangtao Xue, Dongyu Pei, Yu Wang, Hao Zhang, Zherui Cai, Yuchen Gao, Zhiqiang Ma, Ziyuan Hu, Jiakai Gao, Chaochen Gong, Jingjing Song, Yuxuan Zhang, Shuyi Zheng, Xiaoqing Xiong, Deyi Bai, Lei Ouyang, Wanli Zhang, Ya-Qin Ma, Wei-Ying Zhou, Bowen Zhou, Hao |
| contents | We introduce AMix-1, a powerful protein foundation model built on Bayesian Flow Networks and empowered by a systematic training methodology, encompassing pretraining scaling laws, emergent capability analysis, in-context learning mechanism, and test-time scaling algorithm. To guarantee robust scalability, we establish a predictive scaling law and reveal the progressive emergence of structural understanding via loss perspective, culminating in a strong 1.7-billion model. Building on this foundation, we devise a multiple sequence alignment (MSA)-based in-context learning strategy to unify protein design into a general framework, where AMix-1 recognizes deep evolutionary signals among MSAs and consistently generates structurally and functionally coherent proteins. This framework enables the successful design of a dramatically improved AmeR variant with an up to $50\times$ activity increase over its wild type. Pushing the boundaries of protein engineering, we further empower AMix-1 with an evolutionary test-time scaling algorithm for in silico directed evolution that delivers substantial, scalable performance gains as verification budgets are intensified, laying the groundwork for next-generation lab-in-the-loop protein design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_08920 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model Lv, Changze Zhou, Jiang Long, Siyu Wang, Lihao Feng, Jiangtao Xue, Dongyu Pei, Yu Wang, Hao Zhang, Zherui Cai, Yuchen Gao, Zhiqiang Ma, Ziyuan Hu, Jiakai Gao, Chaochen Gong, Jingjing Song, Yuxuan Zhang, Shuyi Zheng, Xiaoqing Xiong, Deyi Bai, Lei Ouyang, Wanli Zhang, Ya-Qin Ma, Wei-Ying Zhou, Bowen Zhou, Hao Biomolecules Artificial Intelligence We introduce AMix-1, a powerful protein foundation model built on Bayesian Flow Networks and empowered by a systematic training methodology, encompassing pretraining scaling laws, emergent capability analysis, in-context learning mechanism, and test-time scaling algorithm. To guarantee robust scalability, we establish a predictive scaling law and reveal the progressive emergence of structural understanding via loss perspective, culminating in a strong 1.7-billion model. Building on this foundation, we devise a multiple sequence alignment (MSA)-based in-context learning strategy to unify protein design into a general framework, where AMix-1 recognizes deep evolutionary signals among MSAs and consistently generates structurally and functionally coherent proteins. This framework enables the successful design of a dramatically improved AmeR variant with an up to $50\times$ activity increase over its wild type. Pushing the boundaries of protein engineering, we further empower AMix-1 with an evolutionary test-time scaling algorithm for in silico directed evolution that delivers substantial, scalable performance gains as verification budgets are intensified, laying the groundwork for next-generation lab-in-the-loop protein design. |
| title | AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model |
| topic | Biomolecules Artificial Intelligence |
| url | https://arxiv.org/abs/2507.08920 |